Correlation model selection for internal target movement

• US 8,180,432 B2
• Filed: 09/29/2005
• Issued: 05/15/2012
• Est. Priority Date: 10/02/2004
• Status: Active Grant

• Current Assignee: Accuray Incorporated
• Original Assignee: Accuray Incorporated
• Inventors: Sayeh, Sohail
• Primary Examiner(s): JAWORSKI, FRANCIS J
• Application Number:
  US11/240,593
• Publication Number:
  US 20060074304A1
• Time in Patent Office: 2,420 Days
• Field of Search: 600/411, 600/413, 600426-428, 606/130, 378/62, 378/69, 382/132
• US Class Current: 600/426
• CPC Class Codes: A61B 5/1127 using markers
• CPC Class Codes: A61B 5/113 occurring during breathing
• CPC Class Codes: A61B 6/4458 the source unit or the detector unit being attached to robotic arms
• CPC Class Codes: A61N 2005/1061 using an x-ray imaging system having a separate imaging source
• CPC Class Codes: A61N 5/1037 taking into account the movement of the target, e.g. 4D-image based planning
• CPC Class Codes: A61N 5/1049 for verifying the position of the patient with respect to the radiation beam
• CPC Class Codes: A61N 5/1064 for adjusting radiation treatment in response to monitoring
• CPC Class Codes: A61N 5/1067 in real time, i.e. during treatment
• CPC Class Codes: A61N 5/107 in real time, i.e. during treatment

Abstract

A method and apparatus to select a correlation model from a plurality of correlation models to track target movement based on movement of an external marker.

Claims

1. A method, implemented by a computing system programmed to perform operations, comprising:

receiving, at the computing system, a plurality of data points representative of a corresponding plurality of positions over time of an external marker associated with a patient;

receiving, at the computing system, a plurality of images of a target within the patient, the plurality of images corresponding to the plurality of data points;

generating a plurality of correlation models using the plurality of images and the plurality of data points, each of the plurality of correlation models descriptive of an internal path of movement over time of the target within the patient as a function of the movement of the external marker;

selecting one of the plurality of correlation models; and

deriving a target position of the target based on the selected correlation model.

2. The method of claim 1, wherein the plurality of correlation models comprises a linear model or a non-linear model.

3. The method of claim 2, wherein the non-linear model comprises at least one second order polynomial approximation to approximate the internal path of movement of the target.

4. The method of claim 2, wherein the non-linear model comprises at least one third order polynomial approximation to approximate the internal path of movement of the target.

5. The method of claim 1, wherein the plurality of correlation models comprises a non-linear model inclusive of a non-linear approximation to approximate the internal path of movement of the target.

6. The method of claim 1, wherein the plurality of correlation models comprises a multi-linear model inclusive of a plurality of linear approximations to approximate the internal path of movement of the target, wherein the plurality of linear approximations comprise a first linear approximation connected to a second linear approximation at a non-zero angle.

7. The method of claim 1, wherein the plurality of correlation models comprises a linear model inclusive of a linear approximation to approximate the internal path of movement of the target.

8. The method of claim 1, further comprising generating a plurality of estimation errors corresponding to the plurality of correlation models, each estimation error descriptive of an approximate error distance between the target position and an actual location of the target.

9. The method of claim 8, wherein the selecting of one of the plurality of correlation models is based on the plurality of estimation errors.

10. The method of claim 8, further comprising assigning a weighting factor to each of the plurality of correlation models, wherein the generating of the plurality of estimation errors is based on the weighting factor corresponding to each of the plurality of correlation models.

11. The method of claim 10, wherein the selecting of one of the plurality of correlation models comprises comparing the plurality of estimation errors with one another and selecting the one of the plurality of correlation models having the lowest estimation error at one of a plurality of displacement points.

12. The method of claim 1, further comprising:

sending a position signal associated with the target position to a beam generator controller; and

controlling a beam generator to direct a beam at the target.

13. The method of claim 1, further comprising:

deselecting the selected correlation model, wherein a first estimation error is associated with the selected correlation model; and

selecting another one of the plurality of correlation models in response to a determination that a second estimation error associated with the other one of the plurality of correlation models is lower than the first estimation error.

14. A non-transitory machine readable storage medium having instructions thereon, which instructions, when executed by a digital processing device, cause the digital processing device to perform the method of claim 1.

15. The non-transitory machine readable storage medium of claim 14, wherein the plurality of correlation models comprises a linear model or a non-linear model, and wherein the non-linear model comprises at least one second order polynomial approximation to approximate the internal path of movement of the target.

16. The non-transitory machine readable storage medium of claim 14, wherein the plurality of correlation models comprises a linear model or a non-linear model, and wherein the non-linear model comprises at least one third order polynomial approximation to approximate the internal path of movement of the target.

17. An apparatus, comprising:

a treatment planning system coupled to a diagnostic imaging system and a treatment delivery system; and

a digital processing device coupled to the treatment planning system, the digital processing device configured to receive a plurality of data points representative of a corresponding plurality of positions over time of an external marker associated with a patient, to receive a plurality of images of a target within the patient, the plurality of images corresponding to the plurality of data points, the digital processing device to generate a plurality of correlation models using the plurality of images and the plurality of data points, each of the plurality of correlation models descriptive of an internal path of movement of the target as a function of the movement of the external marker, and the digital processing device to select one of the plurality of correlation models to derive a target position of the target.

18. The apparatus of claim 17, wherein the plurality of correlation models comprises a linear model, a multi-poly model, or a multi-linear model.

19. The apparatus of claim 18, wherein the multi-poly model comprises at least one second order polynomial approximation to approximate the internal path of movement of the target.

20. The apparatus of claim 18, wherein the multi-poly model comprises at least one third order polynomial approximation to approximate the internal path of movement of the target.

21. The apparatus of claim 17, wherein the digital processing device is further configured to generate a plurality of estimation errors corresponding to the plurality of correlation models and to select one of the plurality of correlation models based on the plurality of estimation errors.

22. The apparatus of claim 21, wherein the digital processing device is further configured to compare the plurality of estimation errors with one another and to select the one of the plurality of correlation models having the lowest estimation error at one of a plurality of displacement points.

23. The apparatus of claim 17, wherein the digital processing device is further configured to assign a weighting factor to each of the plurality of correlation models and to generate the plurality of estimation errors based on the weighting factor corresponding to each of the plurality of correlation models.

24. The apparatus of claim 17, wherein the digital processing device is further configured to deselect the selected correlation model, wherein a first estimation error is associated with the selected correlation model, and to select another one of the plurality of correlation models in response to a determination that a second estimation error associated with the other one of the plurality of correlation models is lower than the first estimation error.

25. An apparatus, comprising:

means for acquiring a plurality of images of a target within a patient;

means for acquiring a plurality of data points of an external marker associated with the patient;

means for generating a plurality of correlation models using the plurality of images of the target and the plurality of data points, each of the plurality of correlation models descriptive of a path of movement over time of the target as a function of the movement of the external marker;

means for determining which one of the plurality of correlation models most closely approximates an actual position of the target at a specified displacement point; and

means for tracking the target using the one of the plurality of correlation models that most closely approximates the actual position of the target.

26. The apparatus of claim 25, further comprising:

means for selecting one of the plurality of correlation models; and

means for deriving a target position of the target based on the selected correlation model.

27. The apparatus of claim 25, further comprising means for generating a plurality of estimation errors corresponding to the plurality of correlation models, each estimation error indicative of an estimated difference between a displacement point on one of the plurality of correlation models and an actual position of the target.